Face Shape Research Articles

Facial soft-tissue shape changes after fixed edgewise treatment with premolar extraction in individual artificial-intelligence-classified facial profile patterns

ObjectiveTo examine the patterns of pretreatment facial soft tissue shape in orthodontic cases with premolar extraction using artificial intelligence (AI) and to investigate the corresponding changes.MethodsOne hundred and fifty-two patients who underwent orthodontic treatment with premolar extraction were enrolled. Lateral cephalograms were obtained before and after the treatment. For each record, the outlines of the nose-lip-chin profile and corresponding 21 cephalometric variables were extracted. The AI method classified pretreatment records into three subject groups based on the feature variables extracted from the outline. Dentoskeletal and soft tissue facial form changes observed after treatment were compared statistically (P < 0.05) between the groups using ANOVA. Multivariate regression models were used for each group.ResultsGroup 1 (n = 59) was characterized by Class II high-angle retrognathic mandible with an incompetent lip, group 2 (n = 55) by Class I malocclusion with retruded and thin lips, and group 3 (n = 38) by Class I malocclusion with an everted superior lip before treatment. The ratios of anteroposterior soft tissue to hard tissue movements in Group 1 were 56% (r = 0.64) and 83% (r = 0.75) for the superior and inferior lips, respectively, whereas those in Group 2 were 49% (r = 0.78) and 91% (r = 0.80), and 40% (r = 0.54) and 79% (r = 0.70), respectively, in Group 3.ConclusionsThe modes of facial form changes differed depending on the pre-treatment profile patterns classified by the AI. This indicates that the determination of the pre-treatment profile pattern can help in the selection of soft tissue to hard tissue movement ratios, which helps estimate the post-treatment facial profile with a moderate to high correlation.

Face-brain correlates as potential sex-specific biomarkers for schizophrenia and bipolar disorder

Given the shared ectodermal origin and integrated development of the face and the brain, facial biomarkers emerge as potential candidates to assess vulnerability for disorders in which neurodevelopment is compromised, such as schizophrenia (SZ) and bipolar disorder (BD).The sample comprised 188 individuals (67 SZ patients, 46 BD patients and 75 healthy controls (HC)). Using a landmark-based approach on 3D facial reconstructions, we quantified global and local facial shape differences between SZ/BD patients and HC using geometric morphometrics. We also assessed correlations between facial and brain cortical measures. All analyses were performed separately by sex.Diagnosis explained 4.1 % - 5.9 % of global facial shape variance in males and females with SZ, and 4.5 % - 4.1 % in BD. Regarding local facial shape, we detected 43.2 % of significantly different distances in males and 47.4 % in females with SZ as compared to HC, whereas in BD the percentages decreased to 35.8 % and 26.8 %, respectively. We detected that brain area and volume significantly explained 2.2 % and 2 % of facial shape variance in the male SZ - HC sample.Our results support facial shape as a neurodevelopmental marker for SZ and BD and reveal sex-specific pathophysiological mechanisms modulating the interplay between the brain and the face.

Critical Face Pressure of a Tunnel Driven by a Shield Machine Considering Seismic Forces and Tunnel Shape Influence

Evaluating critical face pressure with high accuracy is an important topic for shield tunneling. The existing research focuses more on the influence of complex geological conditions or environmental factors, and there are few reports on the influence of tunnel shape on critical face pressure. In reality, the tunnel shape has a significant impact on the deformation zone in front of the tunnel face, which may lead to non-negligible differences in shield pressure. To fill this gap, an efficient analytical approach is proposed to estimate the critical face pressure considering different tunnel shapes in the framework of limit analysis. As earthquakes are potential threats to the tunnel face stability in seismically active regions, the seismic effect is also taken into account with the help of the pseudo-static method. Several cases with the same area but different tunnel shapes are investigated using the limit analysis method. A comparison with static analysis is given to highlight the influence of seismic forces on the tunnel face stability. The results show that the critical face pressures increase by 15.3% when the tunnel face shape changes from rectangular to circular, and by 23.5% when the horizontal seismic coefficient varies from 0 to 0.1. A further validation with a 3D finite difference method is performed with respect to four typical tunnel shapes considered in this study. Lastly, several stability charts are provided for a quick estimation of the critical tunnel face pressure subjected to seismic forces. It is concluded that the proposed method can be applied to a tunnel stability assessment of various cross-sections and is highly efficient compared with numerical simulations.

Mapping genes for human face shape: exploration of univariate phenotyping strategies.

Advancements linking variation in the human genome to phenotypes have rapidly evolved in recent decades and have revealed that most human traits are influenced by genetic variants to at least some degree. While many traits, such as stature, are straightforward to acquire and investigate, the multivariate and multipartite nature of facial shape makes quantification more challenging. In this study, we compared the impact of different facial phenotyping approaches on gene mapping outcomes. Our findings suggest that the choice of facial phenotyping method has an impact on apparent trait heritability and the ability to detect genetic association signals. These results offer valuable insights into the importance of phenotyping in genetic investigations, especially when dealing with highly complex morphological traits.

AI-generated faces show lower morphological diversity than real faces do

Some recent studies suggest that artificial intelligence can create realistic human faces subjectively unrecognizable from faces of real people. We have compared static facial photographs of 197 real men with a sample of 200 male faces generated by artificial intelligence to test whether they converge in basic morphological characteristic such as shape variation and bilateral asymmetry. Both datasets depicted standardized faces of European men with a neutral expression. Then we used geometric morphometrics to investigate their facial morphology and calculate the measures of shape variation and asymmetry. We found that the natural faces of real individuals were more variable in their facial shape than the artificially generated faces were. Moreover, the artificially synthesized faces showed lower levels of facial asymmetry than the control group. Despite the rapid development of generative adversarial networks, natural faces are thus still statistically distinguishable from the artificial ones by objective measurements. We recommend the researchers in face perception, that aim to use artificially generated faces as ecologically valid stimuli, to check whether their stimuli morphological variance is comparable with that of natural faces in a target population.

Изучение краниологических серий из некрополей русских городов XI–XVIII вв. методом геометрической морфометрии

Anthropological features of the Russian city population in the 11th – 18th centuries were studied using classical methods of craniometry as well as cranio- and osteoscopy. However, the method of geometric morphometry, which has been actively used since the 1990s and is a recent morphometric tool, has not yet been applied to Russian urban craniological series. We obtained and analyzed by the GM method three-dimensional copies of 225 skulls from thirteen series of the Middle Ages and the Early Modern period from the excavations of the Institute of Archaeology of the Russian Academy of Sciences. Despite the small number of studied samples from Tver and Torzhok, we were able to trace their features and originality relative to other comparable groups and among themselves. Based on historical information, these variations can be explained by political factors. The differences between the political systems of Ancient Rus’ and Russia in the Early Modern period also explain the greater morphological homogeneity of the urban population in the 15th – 18th centuries compared to the 11th –13th centuries. The study of variability in the late urban series and comparison of diachronic samples from Yaroslavl and Pereslavl-Zalessky showed morphological similarity between the inhabitants of the Middle Ages and the Early Modern period, which may indicate the constancy of administrative, trade, and economic ties, as well as the significant contribution of the local rural population to the formation of the anthropological appearance of the townspeople. The revealed greater variability in the female samples compared to the male ones may indicate significant irregularities in the facial skeleton shape of women, which cannot be disclosed using classical craniology data. In general, the obtained results not only confirm many of the conclusions of previous craniological studies of the urban population from Eastern Europe but also make it possible to obtain new data on the degree of homogeneity of the townspeople’s anthropological appearance in the Middle Ages and Early Modern periods.

Deep Learning–Based Facial and Skeletal Transformations for Surgical Planning

The increasing application of virtual surgical planning (VSP) in orthognathic surgery implies a critical need for accurate prediction of facial and skeletal shapes. The craniofacial relationship in patients with dentofacial deformities is still not understood, and transformations between facial and skeletal shapes remain a challenging task due to intricate anatomical structures and nonlinear relationships between the facial soft tissue and bones. In this study, a novel bidirectional 3-dimensional (3D) deep learning framework, named P2P-ConvGC, was developed and validated based on a large-scale data set for accurate subject-specific transformations between facial and skeletal shapes. Specifically, the 2-stage point-sampling strategy was used to generate multiple nonoverlapping point subsets to represent high-resolution facial and skeletal shapes. Facial and skeletal point subsets were separately input into the prediction system to predict the corresponding skeletal and facial point subsets via the skeletal prediction subnetwork and facial prediction subnetwork. For quantitative evaluation, the accuracy was calculated with shape errors and landmark errors between the predicted skeleton or face with corresponding ground truths. The shape error was calculated by comparing the predicted point sets with the ground truths, with P2P-ConvGC outperforming existing state-of-the-art algorithms including P2P-Net, P2P-ASNL, and P2P-Conv. The total landmark errors (Euclidean distances of craniomaxillofacial landmarks) of P2P-ConvGC in the upper skull, mandible, and facial soft tissues were 1.964 ± 0.904 mm, 2.398 ± 1.174 mm, and 2.226 ± 0.774 mm, respectively. Furthermore, the clinical feasibility of the bidirectional model was validated using a clinical cohort. The result demonstrated its prediction ability with average surface deviation errors of 0.895 ± 0.175 mm for facial prediction and 0.906 ± 0.082 mm for skeletal prediction. To conclude, our proposed model achieved good performance on the subject-specific prediction of facial and skeletal shapes and showed clinical application potential in postoperative facial prediction and VSP for orthognathic surgery.

Association between the three-dimensional facial shape and its color in a boundary group of young to middle-aged Asian women

Visual cues strongly influence an individual's self-esteem and have fundamental sociopsychological functions. The color and shape of the face are important information for visual cues and are hypothesized to be correlated with each other. However, few studies have examined these relationships. Thus, this study determined the association between color and shape of the face. For this purpose, we evaluated Chinese women in their 30s and 40s (n = 166). Three-dimensional (3D) image-capture devices that provide shape morphology along with standardized photographs (color information) were used to obtain 3D images of women. The coordinates and red‒green–blue color data on the 3D images were utilized to perform principal component (PC) analysis, and shape and color PCs were generated. A canonical variate analysis was then conducted to check for significant correlations between the shape and color PCs. As a result, 6 significant correlations were found (p < 0.05). In detail, in addition to the physical correlations (i.e., steric faces or faces with protrusion of the cheek showed greater shadows, retrognathism was related to a shadow under the lower lip and vice versa), biological correlations (fatty faces showed greater redness and remarkable marionette lines; faces with age-related sagging showed greater darkness, possibly related to cumulative ultraviolet radiation exposure of the skin; and robust mandibles and supraorbital ridges were related to thick eyebrows) were found. This insight can aid both medical and cosmetic practitioners in comprehending the intricate details conveyed by facial features, thereby facilitating more comprehensive diagnosis and treatment planning, including makeup.

Improved 3D human face reconstruction from 2D images using blended hard edges

This study reports an effective and robust edge-based scheme for the reconstruction of 3D human faces from input of single images, addressing drawbacks of existing methods in case of large face pose angles or noisy input images. Accurate 3D face reconstruction from 2D images is important, as it can enable a wide range of applications, such as face recognition, animations, games and AR/VR systems. Edge features extracted from 2D images contain wealthy and robust 3D geometric information, which were used together with landmarks for face reconstruction purpose. However, the accurate reconstruction of 3D faces from contour features is a challenging task, since traditional edge or contour detection algorithms introduce a great deal of noise, which would adversely affect the reconstruction. This paper reports on the use of a hard-blended face contour feature from a neural network and a Canny edge extractor for face reconstruction. The quantitative results indicate that our method achieves a notable improvement in face reconstruction with a Euclidean distance error of 1.64 mm and a normal vector distance error of 1.27 mm when compared to the ground truth, outperforming both traditional and other deep learning-based methods. These metrics show particularly significant advancements, especially in face shape reconstruction under large pose angles. The method also achieved higher accuracy and robustness on in-the-wild images under conditions of blurring, makeup, occlusion and poor illumination.

Novel Illumination-invariant Face Recognition Approach via Reflectance-luminance and Local Matching Model with Weighted Voting System

In this study, a novel approach has been introduced for face recognition that is unaffected by changes in illumination. This method is based on the reflectance-luminance model and incorporates local matching using a weighted voting technique to eliminate any artifacts present in the retina images. A total of 37 different linear and nonlinear filters were tested, including high pass and low pass filters, to achieve the reflectance component of the image, which remains invariant to changes in illumination. Among these filters, the maximum filter, which is a simple filter with low computational complexity, yielded the best results in extracting the illumination invariants. It was observed that the illumination invariants obtained through this method outperformed other methods such as QI, SQI, and image enhancement techniques in terms of recognition accuracy. Importantly, the proposed method does not require any prior knowledge about the facial shape or illumination conditions and can be applied to individual images independently. Unlike many existing methods, this approach does not rely on multiple images during the training stage and does not require any parameter selection to generate the illumination invariants. To further enhance the robustness of illumination, a weighted voting system was introduced. Certain regions of the image, which may adversely affect the recognition outcome due to poor illumination, occlusion, noise, or lack of distinctive information, were identified using predefined factors such as grayscale mean, image entropy, and mutual information. The proposed method was also compared to other face recognition methods in the presence of occlusions, and it demonstrated promising results outperforming existing methods. The Python algorithm successfully detects obstructed faces, genders, and ages in videos with a face matching accuracy between 80.9% to 96.9% based on proximity.

Management of Cowlick in Female Hairline Correction Surgery.

To investigate the results of two grafting methods to manage CLs in the FHL of women. The present study was performed to investigate the results of two grafting methods to manage cowlicks (CLs) in the frontal hairline of women. Patients were categorized into two groups based on the grafting method: the matching with the original hair direction (MOD) group and the matching with the CL hair direction (MCD) group. Satisfaction ratings were compared between the two groups. Among 80 patients, 77 had a single CL and 3 had a double CL. In the MOD group, 7 of 12 (58.3%) patients underwent a touch-up procedure for density enhancement with an average of 123 follicular units; the average patient and physician satisfaction ratings were 4.84 and 4.60, respectively. In the MCD group, 7 of 68 (10.3%) patients underwent a touch-up procedure for density enhancement around the CL; the average patient and physician satisfaction ratings were 4.64 and 4.50, respectively. In female hairline correction surgery, the surgeon must choose between MOD and MCD by considering the location, shape, and pattern of the CL; the facial shape; and the patient's preference. Delicate graft placement is required to obtain natural-looking results. This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors. www.springer.com/00266 .

SwinDPSR: Dual-Path Face Super-Resolution Network Integrating Swin Transformer

Whether to use face priors in the face super-resolution (FSR) methods is a symmetry problem.Various face priors are used to describe the overall and local face features, making the generation of super-resolution face images expensive and laborious. FSR methods that do not require any prior information tend to focus too much on the local features of the face, ignoring the modeling of global information. To solve this problem, we propose a dual-path facial image super-resolution network (SwinDPSR) fused with Swin Transformer. The network does not require additional face priors, and it learns global face shape and local face components through two independent branches. In addition, the channel attention ECA module is used to aggregate the global and local face information in the above dual-path sub-networks, which can generate corresponding high-quality face images. The results of face super-resolution reconstruction experiments on public face datasets and a real-scene face dataset show that SwinDPSR is superior to previous advanced methods both in terms of visual effects and objective indicators. The reconstruction results are evaluated with four evaluation metrics: peak signal-to-noise ratio (PSNR), structural similarity (SSIM), learned perceptual image patch similarity (LPIPS), and mean perceptual score (MPS).

Craniofacial anomalies in schizophrenia-relevant GFAP.HMOX10-12m mice.

Subtle craniofacial dysmorphology has been reported in schizophrenia patients. This dysmorphology includes midline facial elongation, frontonasal anomalies and a sexually dimorphic deviation from normal directional asymmetry of the face, with male patients showing reduced and female patients showing enhanced facial asymmetry relative to healthy control subjects. GFAP.HMOX10-12m transgenic mice (Mus musculus) that overexpress heme oxygenase-1 in astrocytes recapitulate many schizophrenia-relevant neurochemical, neuropathological and behavioral features. As morphogenesis of the brain, skull and face are highly interrelated, we hypothesized that GFAP.HMOX10-12m mice may exhibit craniofacial anomalies similar to those reported in persons with schizophrenia. We examined craniofacial anatomy in male GFAP.HMOX10-12m mice and wild-type control mice at the early adulthood age of 6-8 months. We used computer vision techniques for the extraction and analysis of mouse head shape parameters from systematically acquired 2D digital images, and confirmed our results with landmark-based geometric morphometrics. We performed skull bone morphometry using digital calipers to take linear distance measurements between known landmarks. Relative to controls, adult male GFAP.HMOX10-12m mice manifested craniofacial dysmorphology including elongation of the nasal bones, alteration of head shape anisotropy and reduction of directional asymmetry in facial shape features. These findings demonstrate that GFAP.HMOX10-12m mice exhibit craniofacial anomalies resembling those described in schizophrenia patients, implicating heme oxygenase-1 in their development. As a preclinical mouse model, GFAP.HMOX10-12m mice provide a novel opportunity for the study of the etiopathogenesis of craniofacial and other anomalies in schizophrenia and related disorders.

Management of Mandibular Condylar Sagittal Fracture Accompanied by Lateral Condylar Crest Defect.

The sagittal fracture of the mandibular condyle can be fixed with absorbable long screws. Absorbable long screws are generally inserted from the lateral crest of the condyle and are as close as possible to the medial pole of the condyle to obtain sufficient retention force. However, in clinical practice, patients with locally comminuted condylar fractures and partial defects in the lateral crest are often encountered. We validated the use of absorbable plates and long screws to fix mandibular condylar fractures with lateral crest defects, and postoperative follow-up showed good results. The preoperative design indicated that if conventional long screws were used, more soft tissue need to be pulled downward to achieve the appropriate drilling angle. If an absorbable plate was used, the degree of downward pulling of soft tissue was smaller, which can better protect the parotid gland tissue and facial nerve. The surgery was performed according to the preoperative design, using an absorbable plate scheme. Postoperative CT confirmed a stable anatomical reduction of condyle. Four-month follow-up showed that the patient's facial shape, occlusion, and mouth opening were all good. Follow-up CT showed good fracture healing. It is feasible to use absorbable plates and long absorbable screws to fix mandibular condylar sagittal fracture accompanied by lateral condylar crest defect.

Coordinate-wise monotonic transformations enable privacy-preserving age estimation with 3D face point cloud.

The human face is a valuable biomarker of aging, but the collection and use of its image raise significant privacy concerns. Here we present an approach for facial data masking that preserves age-related features using coordinate-wise monotonic transformations. We first develop a deep learning model that estimates age directly from non-registered face point clouds with high accuracy and generalizability. We show that the model learns a highly indistinguishable mapping using faces treated with coordinate-wise monotonic transformations, indicating that the relative positioning of facial information is a low-level biomarker of facial aging. Through visual perception tests and computational 3D face verification experiments, we demonstrate that transformed faces are significantly more difficult to perceive for human but not for machines, except when only the face shape information is accessible. Our study leads to a facial data protection guideline that has the potential to broaden public access to face datasets with minimized privacy risks.

Re-evaluating the Role of Partnership-Related Perceptions in Women's Preferences for Men with Masculine Face Shapes.

Many researchers have proposed that women perceive men with masculine face shapes to be less suitable as parents and long-term partners than men with feminine face shapes, causing women to find masculine men more attractive for short-term than long-term relationships. However, recent work shows that results obtained using the type of experimentally manipulated stimuli that were employed in studies presenting evidence for these claims are not necessarily observed when natural (i.e., unmanipulated) face stimuli were used to suggest that the evidence for these claims may need to be revaluated. Consequently, we tested for possible relationships between ratings of natural male faces for parenting- and relationship-related traits and shape masculinity (Study 1) and also tested whether women's preferences for shape masculinity were stronger when natural male faces were rated for short-term relationships than when natural male faces were rated for long-term relationships (Studies 2 and 3). We saw no evidence for either of these predictions, instead finding that men with more attractive faces were perceived to be better parents and better long-term partners. Thus, our findings do not support the widely held view that masculine men are more attractive for short-term relationships because they are perceived to be unlikely to invest time and effort in their romantic partners and offspring.

A keypoints-motion-based landmark transfer method for face reenactment

Face reenactment is a task to transfer the facial pose and expression from one driving face to the source face. It can be done by adding many condition information, such as the landmark, 3D Morphable Model (3DMM). Because the landmark information is easy to get and contains face structure and expression information, many works rely on it. One of the key points for the landmark-based method is to generate landmarks that keep the face shape of the source image while having the same expression and pose as the driving image. Concentrating on this problem, we propose a novel and effective method to transfer the expression and pose based on the motion of the key points between driving images. Besides, we also try to improve the fit ability of the current popular backbone by adding an enhancement module.

Are 3D Face Shapes Expressive Enough for Recognising Continuous Emotions and Action Unit Intensities?

Recognising continuous emotions and action unit (AU) intensities from face videos, requires a spatial and temporal understanding of expression dynamics. Existing works primarily rely on 2D face appearance features to extract such dynamics. This work focuses on a promising alternative based on parametric 3D face alignment models, which disentangle different factors of variation, including expression-induced shape variations. We aim to understand how expressive 3D face shapes are in estimating valence-arousal and AU intensities compared to the state-of-the-art 2D appearance-based models. We benchmark five recent 3D face models: ExpNet, 3DDFA-V2, RingNet, DECA, and EMOCA. In valence-arousal estimation, expression features of 3D face models consistently surpassed previous works and yielded an average concordance correlation of. 745 and. 574 on SEWA and AVEC 2019 CES corpora, respectively. We also study how 3D face shapes performed on AU intensity estimation on BP4D and DISFA datasets, and report that 3D face features were on par with 2D appearance features in recognising AUs 4, 6, 10, 12, and 25, but not the entire set of AUs. To understand this discrepancy, we conduct a correspondence analysis between valence-arousal and AUs, which points out that accurate prediction of valence-arousal may require the knowledge of only a few AUs.

Phenotype Characteristics of Diiti Cattle in the Coastal Region of Tomini Bay-Gorontalo, Indonesia

This study aims to determine the qualitative and quantitative traits of Diiti cattle in the coastal area of Tomini Bay, Gorontalo, Indonesia. A total of 201 phenotypes (qualitative and quantitative) of Diiti cattle were collected, consisting of 138 females and 63 males aged 4-5 years. Qualitative observations were focused on body color, face shape, horn, and dewlap. Quantitative traits were focused on the eight traits that were influenced by genetic and environmental factors. The methods used were descriptive and chi-square analysis. IBM Statistics SPSS 22 was used to analyze the data obtained. Female Diiti cattle had nine body colors, while the males had seven body colors. Based on the front view, male and female Diiti cattle have hexagonal, triangular, and perpendicular facial shapes of 7.94%, 49.20%, 42.68% and 36.95%, 5.80%, 57.25%, respectively. The body sizes of male and female Diiti cattle were different. The body size p-values of male and female Diiti cattle ranged between 0.00-0.063. Diiti cattle have various body colors. Female cattle were reddish white, whitish red, and white, while male cattle were black, brown, and white. Female cattle do not have a hump, and generally have a hexagonal face shape, flat face line, body size, and weight almost the same as Bali cattle, but smaller than PO cattle. Further characterization of Diiti cattle was required as basic information of Diiti cattle as Gorontalo local cattle.

ATM Security based on Face Verification

Abstract: To provide a reliable security solution, this project focuses on creating an ATM security system based on facial recognition using OpenCV, machine learning, and deep learning. The paper examines how facial recognition technology can improve ATM security, offering a non-intrusive and highly accurate method of identity verification. By analyzing unique facial features, such as facial component sizes and shapes, this technology can authenticate users in real-time reliably. The proposed system integrates facial recognition software with existing ATM infrastructure. Users are prompted to look into a camera for facial authentication when approaching the ATM. Upon a successful match, users gain access to ATM functionalities, ensuring a seamless and secure transaction experience. This ATM security system emphasizes robust user verification before granting financial transaction access. By integrating face recognition and personal question verification, the system provides a multilayered security approach, increasing user confidence and deterring unauthorized access attempts.